Filter Forge - Finding offset from repeat value

Sphinx.

Filter Optimizer

Posts: 1750
Filters: 39

I'm trying to figure out how I can offset a checker component exactly one checker square horizontally.

The principle is simple enough, if we have 4x4 checkers (repeat 2), we offset -25 horizontally to get the next checker in the row. I.e. offset value = 100 / (repeat * 2).

The problem arise when I want to control the offset from an intslider that controls the number of repeats in the checker - I can't seem to get the mapping right.

I know that a solution is possible via the Bias curve somehow, but I can't get it right. Any help appreciated smile;)

I've attached a very simple filter that shows the idea.

next checker offset.ffxml

Njyldgarkn init/prepare process x no. of render blocks x no. of passes

Posted: September 30, 2008 5:37 am

Details E-Mail

Kraellin

Kraellin

Posts: 12749
Filters: 99

try a step curve.

If wishes were horses... there'd be a whole lot of horse crap to clean up!

Craig

Posted: September 30, 2008 9:04 am

Details E-Mail

Sphinx.

Filter Optimizer

Posts: 1750
Filters: 39

Quote
Kraellin wrote: try a step curve.

How?

Njyldgarkn init/prepare process x no. of render blocks x no. of passes

Posted: September 30, 2008 9:13 am

Details E-Mail

Kraellin

Kraellin

Posts: 12749
Filters: 99

forget the step curve. i just looked at the filter. but, what i dont get is the problem here. just use the offset. why doesnt that work? are you trying to get the checker offset all in one jump, is that it? rather than just sliding tthe horizontal along and taking many steps?

If wishes were horses... there'd be a whole lot of horse crap to clean up!

Craig

Posted: September 30, 2008 2:08 pm

Details E-Mail

Crapadilla

lvl 52 Filter Weaver and Official "Filter Forge Seer"

Posts: 4365
Filters: 65

Quote
Sphinx. wrote: The problem arise when I want to control the offset from an intslider that controls the number of repeats in the checker - I can't seem to get the mapping right.

Did you take a look at my Unstable Weave filter? I had a similar problem back then: Controlling a Repeat and an Offset value with the same intslider. I used a switch component that came 'preloaded' with the right offsets. The drawback is, you can on use 12 discrete repeats.

The same technique is also inside my good olde Roofing Shingles filter.

--- Crapadilla says: "Damn you, stupid redundant feature requests!" ;)

Posted: September 30, 2008 2:48 pm

Details E-Mail

ThreeDee

Lost in Space

Posts: 1672
Filters: 112

Step curve would work (for example) like this.

next checker offset mod.ffxml

Posted: September 30, 2008 4:31 pm

Details E-Mail

Sphinx.

Filter Optimizer

Posts: 1750
Filters: 39

Ahh great!.. seems to work as it should, though I have a few problems with high cell divisions (repeats)

Cell Interpolate.ffxml

Njyldgarkn init/prepare process x no. of render blocks x no. of passes

Posted: October 1, 2008 4:22 am

Details E-Mail

ThreeDee

Lost in Space

Posts: 1672
Filters: 112

The problem being...?

Posted: October 1, 2008 5:56 am

Details E-Mail

Sphinx.

Filter Optimizer

Posts: 1750
Filters: 39

At a few Cell Division values lines appears (see image), e.g. Cell Division = 40 (disable AA). I'm not sure what is causing it yet...

Njyldgarkn init/prepare process x no. of render blocks x no. of passes

Posted: October 1, 2008 6:25 am

Details E-Mail

ThreeDee

Lost in Space

Posts: 1672
Filters: 112

Okay, I just saw that when opening a file that wasn't 600x600.

I suspect there is an extremely small inaccuracy in the calculation of the curve which causes the offset to be a tiny fraction off and which is somehow exacerbated by the anti-aliasing routine. Seems so, for as you increase the anti-aliasing setting those random lines become less and less visible.

Check out this rather weird way of handling it: Instead of offsetting by 100, put 99.9999 in the offsets and it is gone.

Posted: October 1, 2008 6:49 am

Details E-Mail

Sphinx.

Filter Optimizer

Posts: 1750
Filters: 39